Pharmacology

Cannabis-based medicines: What the human pharmacokinetic data actually shows

An interview with Professor Dr Geert Jan Groeneveld, Centre for Human Drug Research (CHDR), The Netherlands

Cannabis-based medicine is often promoted on a simple narrative: THC gets you high, CBD calms it down, and together they’re safer and more effective than either alone. Professor Geert Jan Groeneveld’s group at the Centre for Human Drug Research (CHDR) in Leiden has spent the last several years testing that theory, producing robust human pharmacokinetic and pharmacodynamic data which counter that – the results don’t support it. The reasons why turn out to matter a great deal for how cannabis-based medicines should be formulated, prescribed, dosed, and monitored.

Q: Your group at CHDR has spent years testing the THC:CBD theory. Does the science support it?
Prof. Groeneveld: No. The pharmacological data does not support it. We’ve run multiple controlled studies looking at THC, CBD, and their combination, and the data consistently contradicts the idea that CBD ‘tempers’ THC.

CBD and CB1: A real mechanism
Q: Let’s start with the receptor-level mechanism. What does CBD do at the CB1 receptor?
Prof. Groeneveld: The mechanism is real, but it comes from Laprairie and colleagues. They showed that CBD doesn’t compete with THC at the CB1 orthosteric site. Instead, CBD binds an allosteric site, and structural biology has since proposed a specific mechanism: CBD binding triggers an outward rotation of the first two transmembrane helices, expanding the orthosteric pocket.

That reduces both the binding affinity and signalling efficacy of THC and the endocannabinoids. Mechanistically, you’d predict CBD should dampen THC’s effects. That’s the basis of the ‘CBD balances THC’ claim. But mechanisms in isolated receptor systems don’t always predict what happens in humans. The human trial that tested the mechanism, and found the opposite

Q: Your group ran the definitive human trial. What did you find?
Prof. Groeneveld: We ran a randomised, double-blind, placebo-controlled, five-way crossover trial in 37 healthy volunteers. We compared THC 9 mg alone with THC plus 10 mg, 30 mg, or 450 mg CBD.

The title of the paper says it plainly: cannabidiol increased psychotropic effects and plasma concentrations of THC, without improving analgesia. At 450 mg CBD, subjective ‘feeling high’ ratings increased by about 60%, along with greater
psychomotor impairment, cognitive disruption, and autonomic effects. CBD didn’t improve THC’s analgesic effect at any dose. That’s the opposite of what the receptor-level mechanism would predict.

Why the mechanism fails in humans: Pharmacokinetics, not pharmacodynamics
Q: So what explains the discrepancy?
Prof. Groeneveld: Pharmacokinetics. CBD inhibits CYP3A4 and CYP2C9, enzymes involved in the metabolism of THC. CBD raises plasma concentrations of THC and its active metabolite 11-OH-THC. Even at 30 mg CBD, we saw evidence of a pharmacokinetic interaction. This suggests interaction risk may emerge at lower doses than many clinicians assume, although the magnitude and clinical relevance will depend on the co-administered medicine and CBD dose.

At higher CBD doses, that PK effect overwhelms any dampening happening at the receptor. It’s a clean illustration of a broader principle: in vitro mechanisms don’t automatically predict in vivo outcomes once you account for metabolism.

Beyond THC: The same PK-first pattern in epilepsy
Q: You’ve also studied CBD in epilepsy. Does the same PK-dominant pattern appear there?
Prof. Groeneveld: Yes. We hypothesised that CBD’s anticonvulsant benefit in Dravet and Lennox-Gastaut syndromes is largely due to a PK interaction with clobazam. CBD raises exposure to clobazam’s active metabolite. We’ve run clinical trial simulations to test this, and we’ve also used TMS/EEG/EMG 1 in healthy volunteers to isolate CBD’s intrinsic CNS effects. A meaningful share of CBDs clinical effect (in pain and epilepsy) appears to come from what it does to other medicines, not solely from its own intrinsic action.

1 TMS: Transcranial Magnetic Stimulation; EEG: Electroencephalography; EMG: Electromyography

The interaction risk that matters for everyday co-prescribing
Q: Your recent work looked at CBD with common analgesics. What did you find?
Prof. Groeneveld: We tested CBD with amitriptyline and tramadol, two drugs frequently used by chronic pain patients. The key finding was simple: low-dose CBD increases plasma concentrations of amitriptyline and of active metabolites of tramadol. Low-dose is the important point. Many clinicians assume interaction risk only appears at high, anticonvulsant-range doses. But we found CYP-mediated interaction at doses well below that.

This matters enormously for chronic pain, oncology, and palliative care. This is exactly the polypharmacy populations most likely to be taking CYP3A4, CYP2C9, or CYP2C19 substrates. Even at low doses, CBD can cause clinically relevant medicine interactions.

CBD’s own PK and toxicity profile
Q: What should clinicians know about CBD’s own pharmacokinetics and toxicity?
Prof. Groeneveld: CBD undergoes extensive first-pass hepatic metabolism, mainly via CYP3A4 and CYP2C19, with UGT pathways (UDP-glucuronosyltransferase enzyme) as secondary clearance. Oral bioavailability is low and variable, and food significantly affects absorption. Because CBD is both a substrate and inhibitor of these enzymes, it sits at the centre of a wide interaction network.

Clinically significant ALT (alanine aminotransferase) and AST (aspartate aminotransferase) elevations have been reported primarily with high-dose prescription cannabidiol, typically in the 10-20 mg/kg/day range used in epilepsy treatment, especially when combined with valproate or clobazam. CBD is generally well tolerated at typical adjunct doses, but dose and medicine combinations matter. That’s why liver function monitoring is standard at higher doses.

What this means for the next generation of cannabis-based medicines
Q: Given all this, what should medicine developers, prescribers and pharmacists take away?
Prof. Groeneveld: A few clear points:

• Don’t market CBD as a THC ‘balancer’ on the strength of receptor pharmacology alone. The mechanism is real; the net clinical effect in humans, at clinically relevant doses, has been shown to run the other way.
• Treat CBD as a CYP-modulating drug. CBD is not an inert or ‘natural’ adjunct. Its interaction profile needs to be built into prescribing guidance for any product that contains it, scaled to the actual dose in that product.
• Assess interaction risk by dose, not by class. Low-dose CBD products are not automatically low-risk for co-medication interactions.

The central message from the CHDR programme is that CBD should be viewed not as an inert companion cannabinoid, but as a pharmacologically active, CYP-modulating active substance whose clinical effects may derive as much from altering the exposure of co-administered medicines as from its own intrinsic activity.

References
Woodbridge M. Interview with Professor Dr Geert Jan Groeneveld, Centre for Human Drug Research (CHDR), Leiden, The Netherlands. Conducted 2025. Unpublished source material.

Laprairie RB, Bagher AM, Kelly MEM, Denovan-Wright EM. (2015). Cannabidiol is a negative allosteric modulator of the cannabinoid CB1 receptor. British Journal of Pharmacology. 172(20):4790-4805. Cannabidiol is a negative allosteric modulator of the cannabinoid CB1 receptor. – Abstract – Europe PMC

Gorbenko A, Heuberger J, Klumpers L, de Kam M, Strugala P, de Visser S, Groeneveld G. (2024). Cannabidiol Increases Psychotropic Effects and Plasma Concentrations of Δ9-Tetrahydrocannabinol Without Improving Its Analgesic Properties. Clinical Pharmacology & Therapeutics. 116(5):1289-1303. Cannabidiol Increases Psychotropic Effects and Plasma Concentrations of Δ9-Tetrahydrocannabinol Without Improving Its Analgesic Properties

Gorbenko A, de Cuba C, de Goede A, Post T, Bohoslavsky R, Strugala P, Heuberger J, Groeneveld G, et al. (2025). Low-dose cannabidiol increases plasma concentrations of amitriptyline: A clinical drug-drug interaction study. British Journal of Clinical Pharmacology. 92(6):1595-1606. Low-dose cannabidiol increases plasma concentrations of amitriptyline: A clinical drug–drug interaction study – Gorbenko -2026 – British Journal of Clinical Pharmacology – Wiley Online Library

Centre for Human Drug Research (CHDR). Clinical pharmacology demythologises the effects of cannabis. FAST. 26 November 2024. Clinical pharmacology demythologises the effects of cannabis – FAST